Too Geeky?

9 views
Skip to first unread message

Aaron Huslage

unread,
Feb 25, 2011, 8:41:27 PM2/25/11
to building-a-distributed...@googlegroups.com
There are some major issues with the technology stuff we talk about. It's largely too geeky. This is normal for new technology since the developers and early adopters are practitioners of geekery, but I worry about things when they get to grandma and grandpa.

"Normal" people don't know how to use BitTorrent or port forward their router, for instance, so how can they participate in one of the many p2p networks in existence today?

"Normal" people don't even know they HAVE a router, they take what the telco gives them and call them when it breaks.

"Normal" people don't even know what a web browser is, they simply click the little Internet icon on their desktop and are whisked away to a world where everything is peaceful and amazing.

See what I mean? What implications does a distributed Internet have for users? How do we make this stuff EASY and as fully baked as possible?

--
Aaron Huslage -  +1-919-600-1712
http://blog.hact.net
IM: AIM - ahuslage; Yahoo - ahuslage; MSN - hus...@gmail.com; GTalk - hus...@gmail.com

Robert Steele

unread,
Feb 25, 2011, 8:52:30 PM2/25/11
to building-a-distributed...@googlegroups.com
Build them strong, disposable, and interoperable.  Have a recylcling plan.  Really inexpensive cell phones that do SMS, solar-powered Internet routers that are also recyclable, multiple satellite, microwave, and fiber paths; redundancy squared.  You guys are all smarter than I am about technical, but I am seeing a strategic architecture and specific open source solutions in all that has been said up to this point.

Aaron Huslage

unread,
Feb 25, 2011, 9:00:33 PM2/25/11
to building-a-distributed...@googlegroups.com, Robert Steele
I'm talking about user experience. Who's gonna want something they have to think about setting up? It has to be idiotproof...more than we have today...before it will be used widely as an alternative to what's already there.

(i agree about recyclability, btw)

Richard Adler

unread,
Feb 25, 2011, 9:27:03 PM2/25/11
to building-a-distributed...@googlegroups.com, Aaron Huslage, Robert Steele
I agree, Aaron, that it does have to get there eventually. (Oh, *how* I agree!)

But it doesn't have to be there on day one, in the early stages when we're working out the bugs.

There's a balance here: on the one hand, always develop with the idea that *someday* the interfaces must be idiotproof. But allow ourselves the time to address all the other challenges of making it work in the first place, or else we'll never get there.

Having said that, a lot of projects never achieve anything remotely like usability because they don't give the interfaces any thought at all until the very end, when remaining energy and resources are basically used up. So it's always a good idea to raise usability early on as a virtue, so everyone has it mind from the start.

Richard

Aaron Huslage

unread,
Feb 25, 2011, 9:32:37 PM2/25/11
to Richard Adler, building-a-distributed...@googlegroups.com, Robert Steele
This does make sense from a programmatic perspective, but should any project get some amount of press you will find that you only get one chance to make a first impression. User Centered Design is probably a good methodology to at least have in the back of your head while building these things...if possible have a UX person on staff or at least reviewing things periodically so that you are on the right track.

Eugen Leitl

unread,
Feb 26, 2011, 2:47:38 PM2/26/11
to building-a-distributed...@googlegroups.com
On Fri, Feb 25, 2011 at 09:00:33PM -0500, Aaron Huslage wrote:
> I'm talking about user experience. Who's gonna want something they have to
> think about setting up? It has to be idiotproof...more than we have

The basic idea is of a not-quite-transparent bridge. A plug
with two NICs would do, or a WiFi plug with a NIC connection
that goes to the ISP's router or modem.

You can use a captive portal for first HTTP connection,
showing gateway into the network. You would offer alternative
namespace extensions like http://en.wikipedia.org/wiki/.onion
or http://www.i2p2.de/naming.html

While the hash names are immune to jamming, they're not
human friendly, so your portal will need a P2P search
engine, e.g. http://yacy.net/en/ so that people can find
content in alternative namespaces. Your biggest problem
will hostile users.

This can be done completely in software, of course, but then
you're facing a support nightmare. Plus, plug computers will
have less volatility. No need to tax the nascent networks
with a node flux.

You'll need a moderately beefy plug computers, and some
local storage, preferably as nonvolatile solid state drive.

> today...before it will be used widely as an alternative to what's already
> there.
>
> (i agree about recyclability, btw)

--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE

Reply all
Reply to author
Forward
0 new messages